Full text loading...
-
Automated Extraction of Depositional Elements and Geological Features From 3D seismic Data
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 81st EAGE Conference and Exhibition 2019, Jun 2019, Volume 2019, p.1 - 5
Abstract
Understanding the representation of depositional elements and geological features in 3D seismic data and extracting these objects is crucial for building comprehensive geologic models used in reservoir characterization. Here, a best-practice procedure for the automated extraction of seismic geomorphology is presented. First, seismic volume attributes are applied on both original and flattened seismic data to visualize and enhance seismic geomorphology. Selected features are then extracted with three different automated methods: (1) a level set method, (2) a non-local and multi-attribute search method referred to as seismic DNA in combination with Extrema technology, and (3) a convolutional neural network (CNN). All automated methodologies produce accurate results while reducing the amount of interactions, thereby being more efficient compared to manual interpretation.